Re: Benchmarking the IPL
Yes, understood. Check with your IBM representative, but financially I don't think you would see an IBM difference between warm and cold standby. So why not warm? - - - - - Timothy Sipples Resident Architect (Based in Singapore) STG Value Creation and Complex Deals Team IBM Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
IBM offers features and pricing options that should fit your needs perfectly. I think the feature/pricing you want is called CBU. HTH and good luck. -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Kelman, Tom Sent: Friday, May 14, 2010 9:13 AM To: IBM-MAIN@bama.ua.edu Subject: Benchmarking the IPL My management wants to benchmark the IPL process to see how long it will take at various levels of MIPS, MSUs, or whatever measurement criteria can be used. The purpose of this is to determine what the smallest z10 is we can contract for our DR site that will IPL in a reasonable time frame. The plans are to contract for the smallest system possible with the provisions of being able to bump it up during the few times each year we do a full blown DR test or in the event the is an actual disaster situation. So we might contract for a 2098-A03 and the bump it all the way to a 2098-Z03 when needed. Has anyone ever done something similar to this? Is there any way to set a cap on the system that will take affect immediately and be active during the IPL process? I know that the standard softcap won't work since that doesn't take affect until the MSU four hour rolling average hits the cap, which wouldn't happen during an IPL. What I know as a hard cap involves restricting the LPAR to its weight, so I don't think that would work either. Is there any other way to cap a system? Tom Kelman Enterprise Capacity Planner Commerce Bank of Kansas City (816) 760-7632 NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
Timothy, I saw your post talking about having an A01 warm standby with Z03 CBU. That's similar to what my management is thinking, but without the warm standby. We have EMC storage and it we do synchronous mirroring from our main site to our DR site. Since that is all done via the EMC control units we don't need to have the system up for it. We do want to go to the DR site as needed to do minimal connectivity tests. For those we plan to just IPL the A01 system. Then, when we need to do serious DR tests we would activate the Z03 and IPL that. Basically, the whole purpose of this exercise is to determine how long it would take to be up and ready for those minimal connectivity tests. Tom Kelman Enterprise Capacity Planner Commerce Bank of Kansas City (816) 760-7632 -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Timothy Sipples Sent: Sunday, May 16, 2010 5:26 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Benchmarking the IPL Tom, I'm a little confused by your question. Reading between the lines, you seem to be describing IBM's standard Capacity Backup (CBU) offering as part of your contract. CBU covers both actual disasters and a certain number of rehearsals over the contract period. There are no additional IBM software charges (beyond your normal charges) associated with CBU activations, provided you're adhering to the terms of the CBU contract. What confuses me is why you'd be IPLing outside the CBU terms. Everything you described seems like it would be covered as part of CBU, and if you have a CBU capacity of Z03 then you'll be experiencing IPL times associated with the Z03 capacity (upon CBU activation). Could you elaborate a bit on what scenario(s) you have in mind? Thanks. - - - - - Timothy Sipples Resident Architect (Based in Singapore) STG Value Creation and Complex Deals Team IBM Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html * If you wish to communicate securely with Commerce Bank and its affiliates, you must log into your account under Online Services at http://www.commercebank.com or use the Commerce Bank Secure Email Message Center at https://securemail.commercebank.com NOTICE: This electronic mail message and any attached files are confidential. The information is exclusively for the use of the individual or entity intended as the recipient. If you are not the intended recipient, any use, copying, printing, reviewing, retention, disclosure, distribution or forwarding of the message or any attached file is not authorized and is strictly prohibited. If you have received this electronic mail message in error, please advise the sender by reply electronic mail immediately and permanently delete the original transmission, any attachments and any copies of this message from your computer system. * -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
On Mon, 17 May 2010 07:41:33 -0500, Kelman, Tom thomas.kel...@commercebank.com wrote: I saw your post talking about having an A01 warm standby with Z03 CBU. Tom, Thats the setup we are using: z9BC A01 at an alternate site. FWIW, our most recent IPLs of z/OS 1.11 on that machine took approxmiately 7 minutes from start to 'TCAS ACCEPTING LOGONS', keeping in mind that time includes a few manual replies such as '1,noreq' etc. Regards Dana -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
Tom, if you're running as a guest under z/vm, which some D/R vendors do, then you could cap it with a set share xxx absolute limithard. You didn't say if this was your box or a D/R Vendor box. If there's no z/VM, then LPAR weights is what you have to work with. I would definitely check out the referenced redbook. And for a more definitive IPL time, check out IPLDATA STATUS in any dump in IPCS, it will show you IPL and NIP times. Mary Anne -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
Tom, z/OS contains a cblock IPST pointed by ECVTIPST. This cblock contains information about the IPL time for various steps. SHOWzOS display the data and there is also a IPCS command but can't remember the name. Roland Timothy, I saw your post talking about having an A01 warm standby with Z03 CBU. That's similar to what my management is thinking, but without the warm standby. We have EMC storage and it we do synchronous mirroring from our main site to our DR site. Since that is all done via the EMC control units we don't need to have the system up for it. We do want to go to the DR site as needed to do minimal connectivity tests. For those we plan to just IPL the A01 system. Then, when we need to do serious DR tests we would activate the Z03 and IPL that. Basically, the whole purpose of this exercise is to determine how long it would take to be up and ready for those minimal connectivity tests. Tom Kelman Enterprise Capacity Planner Commerce Bank of Kansas City (816) 760-7632 -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Timothy Sipples Sent: Sunday, May 16, 2010 5:26 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Benchmarking the IPL Tom, I'm a little confused by your question. Reading between the lines, you seem to be describing IBM's standard Capacity Backup (CBU) offering as part of your contract. CBU covers both actual disasters and a certain number of rehearsals over the contract period. There are no additional IBM software charges (beyond your normal charges) associated with CBU activations, provided you're adhering to the terms of the CBU contract. What confuses me is why you'd be IPLing outside the CBU terms. Everything you described seems like it would be covered as part of CBU, and if you have a CBU capacity of Z03 then you'll be experiencing IPL times associated with the Z03 capacity (upon CBU activation). Could you elaborate a bit on what scenario(s) you have in mind? Thanks. - - - - - Timothy Sipples Resident Architect (Based in Singapore) STG Value Creation and Complex Deals Team IBM Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html ** *** If you wish to communicate securely with Commerce Bank and its affiliates, you must log into your account under Online Services at http://www.commercebank.com or use the Commerce Bank Secure Email Message Center at https://securemail.commercebank.com NOTICE: This electronic mail message and any attached files are confidential. The information is exclusively for the use of the individual or entity intended as the recipient. If you are not the intended recipient, any use, copying, printing, reviewing, retention, disclosure, distribution or forwarding of the message or any attached file is not authorized and is strictly prohibited. If you have received this electronic mail message in error, please advise the sender by reply electronic mail immediately and permanently delete the original transmission, any attachments and any copies of this message from your computer system. ** *** -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
On Mon, 17 May 2010 09:38:27 -0500, Roland Schiradin rol...@schiradin.de wrote: Tom, z/OS contains a cblock IPST pointed by ECVTIPST. This cblock contains information about the IPL time for various steps. SHOWzOS display the data and there is also a IPCS command but can't remember the name. Roland IPLDATA STATUS against ACTIVE storage. Regards, Art Gutowski Ford Motor Company -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
Tom, I'm a little confused by your question. Reading between the lines, you seem to be describing IBM's standard Capacity Backup (CBU) offering as part of your contract. CBU covers both actual disasters and a certain number of rehearsals over the contract period. There are no additional IBM software charges (beyond your normal charges) associated with CBU activations, provided you're adhering to the terms of the CBU contract. What confuses me is why you'd be IPLing outside the CBU terms. Everything you described seems like it would be covered as part of CBU, and if you have a CBU capacity of Z03 then you'll be experiencing IPL times associated with the Z03 capacity (upon CBU activation). Could you elaborate a bit on what scenario(s) you have in mind? Thanks. - - - - - Timothy Sipples Resident Architect (Based in Singapore) STG Value Creation and Complex Deals Team IBM Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
As a general statement (IPL is primarily I/O-bound), I agree, especially when considering only z/OS. But over time that's probably becoming less true. For example, WebSphere Application Server (on any platform) has certain CPU demands (which can include zAAPs) to get going, especially with certain applications, although there's been quite a bit of progress on that front recently. It's probably also worth pointing out that many shops have significantly reduced the frequency of IPLs, and in particular they've designed their overall environment to avoid IPLs when there could be user demands pending. I'm referring in particular to such things as a warm standby production LPAR, Parallel Sysplex, and GDPS, as examples. For example, having an LPAR on warm standby at a remote site on a BC machine (typically A01 capacity plus CBU) is a very popular (and affordable) choice among shops that don't have GDPS. IPL time doesn't factor into the recovery time in that particular design. - - - - - Timothy Sipples Resident Architect (Based in Singapore) STG Value Creation and Complex Deals Team IBM Growth Markets E-Mail: timothy.sipp...@us.ibm.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Benchmarking the IPL
My management wants to benchmark the IPL process to see how long it will take at various levels of MIPS, MSUs, or whatever measurement criteria can be used. The purpose of this is to determine what the smallest z10 is we can contract for our DR site that will IPL in a reasonable time frame. The plans are to contract for the smallest system possible with the provisions of being able to bump it up during the few times each year we do a full blown DR test or in the event the is an actual disaster situation. So we might contract for a 2098-A03 and the bump it all the way to a 2098-Z03 when needed. Has anyone ever done something similar to this? Is there any way to set a cap on the system that will take affect immediately and be active during the IPL process? I know that the standard softcap won't work since that doesn't take affect until the MSU four hour rolling average hits the cap, which wouldn't happen during an IPL. What I know as a hard cap involves restricting the LPAR to its weight, so I don't think that would work either. Is there any other way to cap a system? Tom Kelman Enterprise Capacity Planner Commerce Bank of Kansas City (816) 760-7632 * If you wish to communicate securely with Commerce Bank and its affiliates, you must log into your account under Online Services at http://www.commercebank.com or use the Commerce Bank Secure Email Message Center at https://securemail.commercebank.com NOTICE: This electronic mail message and any attached files are confidential. The information is exclusively for the use of the individual or entity intended as the recipient. If you are not the intended recipient, any use, copying, printing, reviewing, retention, disclosure, distribution or forwarding of the message or any attached file is not authorized and is strictly prohibited. If you have received this electronic mail message in error, please advise the sender by reply electronic mail immediately and permanently delete the original transmission, any attachments and any copies of this message from your computer system. * -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
Tom - General numbers here, but our 449 MIP z9 2096-S02 IPLs in about 5 minutes and the new z10 CBU machine we have at 26 mips, a 2098-A01 when its not turned up, manages to still IPL the same processes in 15 minutes or so. Hope that helps. Say.. if you have a second... can you share where your data centers are? I'm debating with management right now on the distance between our data centers being too close and I'm collecting information from anyone I can talk to about how far apart others have their's. Thanks. Jeffrey Deaver, Engineer Systems Engineering jeffrey.dea...@securian.com 651-665-4231(v) IS - Creating competitive advantage with technology. Providing service that excels. OSS - Where Innovation Happens Kelman, Tom thomas.kel...@co MMERCEBANK.COMTo Sent by: IBM IBM-MAIN@bama.ua.edu Mainframe cc Discussion List ibm-m...@bama.ua Subject .edu Benchmarking the IPL 05/14/2010 09:13 AM Please respond to IBM Mainframe Discussion List ibm-m...@bama.ua .edu My management wants to benchmark the IPL process to see how long it will take at various levels of MIPS, MSUs, or whatever measurement criteria can be used. The purpose of this is to determine what the smallest z10 is we can contract for our DR site that will IPL in a reasonable time frame. The plans are to contract for the smallest system possible with the provisions of being able to bump it up during the few times each year we do a full blown DR test or in the event the is an actual disaster situation. So we might contract for a 2098-A03 and the bump it all the way to a 2098-Z03 when needed. Has anyone ever done something similar to this? Is there any way to set a cap on the system that will take affect immediately and be active during the IPL process? I know that the standard softcap won't work since that doesn't take affect until the MSU four hour rolling average hits the cap, which wouldn't happen during an IPL. What I know as a hard cap involves restricting the LPAR to its weight, so I don't think that would work either. Is there any other way to cap a system? Tom Kelman Enterprise Capacity Planner Commerce Bank of Kansas City (816) 760-7632 * If you wish to communicate securely with Commerce Bank and its affiliates, you must log into your account under Online Services at http://www.commercebank.com or use the Commerce Bank Secure Email Message Center at https://securemail.commercebank.com NOTICE: This electronic mail message and any attached files are confidential. The information is exclusively for the use of the individual or entity intended as the recipient. If you are not the intended recipient, any use, copying, printing, reviewing, retention, disclosure, distribution or forwarding of the message or any attached file is not authorized and is strictly prohibited. If you have received this electronic mail message in error, please advise the sender by reply electronic mail immediately and permanently delete the original transmission, any attachments and any copies of this message from your computer system. * -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe
Re: Benchmarking the IPL
Obviously, that was an email forwarding mistake. I'm NOT looking for a mass response on data center distances. Thanks. Jeffrey Deaver, Engineer Systems Engineering jeffrey.dea...@securian.com 651-665-4231(v) IS - Creating competitive advantage with technology. Providing service that excels. OSS - Where Innovation Happens Jeffrey Deaver jeffrey.dea...@s ECURIAN.COM To Sent by: IBM IBM-MAIN@bama.ua.edu Mainframe cc Discussion List ibm-m...@bama.ua Subject .edu Re: Benchmarking the IPL 05/14/2010 10:10 AM Please respond to IBM Mainframe Discussion List ibm-m...@bama.ua .edu Tom - General numbers here, but our 449 MIP z9 2096-S02 IPLs in about 5 minutes and the new z10 CBU machine we have at 26 mips, a 2098-A01 when its not turned up, manages to still IPL the same processes in 15 minutes or so. Hope that helps. Say.. if you have a second... can you share where your data centers are? I'm debating with management right now on the distance between our data centers being too close and I'm collecting information from anyone I can talk to about how far apart others have their's. Thanks. Jeffrey Deaver, Engineer Systems Engineering jeffrey.dea...@securian.com 651-665-4231(v) IS - Creating competitive advantage with technology. Providing service that excels. OSS - Where Innovation Happens Kelman, Tom thomas.kel...@co MMERCEBANK.COMTo Sent by: IBM IBM-MAIN@bama.ua.edu Mainframe cc Discussion List ibm-m...@bama.ua Subject .edu Benchmarking the IPL 05/14/2010 09:13 AM Please respond to IBM Mainframe Discussion List ibm-m...@bama.ua .edu My management wants to benchmark the IPL process to see how long it will take at various levels of MIPS, MSUs, or whatever measurement criteria can be used. The purpose of this is to determine what the smallest z10 is we can contract for our DR site that will IPL in a reasonable time frame. The plans are to contract for the smallest system possible with the provisions of being able to bump it up during the few times each year we do a full blown DR test or in the event the is an actual disaster situation. So we might contract for a 2098-A03 and the bump it all the way to a 2098-Z03 when needed. Has anyone ever done something similar to this? Is there any way to set a cap on the system that will take affect immediately and be active during the IPL process? I know that the standard softcap won't work since that doesn't take affect until the MSU four hour rolling average hits the cap, which wouldn't happen during an IPL. What I know as a hard cap involves restricting the LPAR to its weight, so I don't think that would work either. Is there any other way to cap a system? Tom Kelman Enterprise Capacity Planner Commerce Bank of Kansas City (816) 760-7632 * If you wish to communicate securely with Commerce Bank and its affiliates, you must log into your account under Online Services at http://www.commercebank.com or use the Commerce Bank Secure Email Message Center at https://securemail.commercebank.com NOTICE: This electronic mail message and any attached files are confidential. The information is exclusively
Re: Benchmarking the IPL
Jeffrey Deaver wrote: I'm NOT looking for a mass response on data center distances. Only one mass response from me: If there is a fence between the two data centers, then they are too NEAR !!! ;-D Is it already Friday? Groete / Greetings Elardus Engelbrecht -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
I'm debating with management right now on the distance between our data centers being too close and I'm collecting information from anyone I can talk to about how far apart others have their's. I've worked at two shops, and they were at (potential) extremes. One had the primary just north of Toronto, and the second just above the lake shore. This was due to the original restrictions of ESCON extension limitations. The second had the primary in Dallas, and the other in Philidelphia. Recent disasters, like Katrina, make me believe you should be as far away as you can afford. And, I would use examples like Katrina in my discussions with Management. - Too busy driving to stop for gas! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Benchmarking the IPL
Greetings, Generally speaking as I understand it CPU capacity is not a major factor in the duration of an IPL. It is mostly IO bound. Have you checked out the following somewhat recent IBM Redbook? System z Mean Time to Recovery Best Practices http://www.redbooks.ibm.com/abstracts/sg247816.html?Open It has a myriad of useful information, one being IPL durations. It presents insightful things such as: - A methodology to measure the details of IPLs - IPL duration improvements over time at different z/OS levels - Parameters that can elongate IPLs A high level summary could be that getting current on z/OS levels, SMF Type 19 Record collection and fast channels/DASD are the major factors. Kudos to the Redbook Authors: Frank Kyne Judi Bank David Sanders Mark Todd David Viguers Cheryl Watson Shulian Yang Cheers... Michael -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
